SUMMARY New treatment approaches to traumatic brain injury (TBI) are greatly needed. The annual economic cost of TBI has been estimated to be $76.5 billion. Though a variety of medical interventions can be implemented in the acute phase, many patients with mild to moderate TBI (mmTBI) remain symptomatic months and years after injury. Commonly available treatments are typically limited to conventional rehabilitation strategies, characterized by only modest efficacy. We propose to investigate transcranial direct current stimulation (tDCS) as a treatment for the cognitive and emotional difficulties commonly observed by clinicians in the post-acute period. Our research team has obtained striking data revealing that tDCS increases the rate of learning in healthy controls. Other researchers have shown that tDCS can facilitate recovery in chronic stroke patients with aphasia. As tDCS is safe, economical, and has the potential for widespread implementation in diverse health facilities, demonstration of clinical efficacy in mmTBI would represent a tremendous step forward for patients and their families. Several important features define the specific approach adopted here. First, we are interested in improving both cognitive function and mood, as these two types of symptoms occur in most patients. Given this goal, we are choosing to implement a tDCS protocol based on left prefrontal anodal stimulation. Though we believe this protocol provides the best chance for clinical success, we will carefully evaluate clinical predictors of outcome (such as severity of injury and premorbid intelligence). Second, we are interested in the persistence of any tDCS gains. Hence, we have designed both short-term (one month post- tDCS) and long-term (six month and one year) follow-ups. Third, our study design is based on an explicit tripartite model of prefrontal function that guided the development of NIH's Examiner battery of cognitive control tasks, which will serve as primary outcome measures. The neural circuitry underlying control overlaps with that of emotion regulation, suggesting that a single intervention may improve both domains. Fourth, our data analysis strategy (linear mixed models) provides longitudinal analyses minimally impacted by the types of missing data problems endemic to longitudinal clinical studies. These design features allow emergent synergy with other components of the Center, especially Cavanaugh's cognitive control/EEG study and Bragin's rodent TBI model/tDCS study. This study will also benefit greatly from the Clinical Core, which will provide equipment, neurocognitive assessments, patient recruitment, and mentoring for subsequent grant submissions.